CPS commission settlement tax calculation method and device, computer equipment and storage medium

Abstract

According to the CPS commission settlement tax calculation method provided by the invention, the user portrait is generated according to the target user information, the user features can be accurately described, and a data basis is provided for subsequent tax calculation rule matching; and the tax calculation rule to be selected is screened in the tax calculation rule base based on the user portrait, so that accurate association between the tax calculation rule and the actual condition of the user is realized. And the scheme income of each to-be-selected tax calculation rule is evaluated and the optimal rule is selected, so that the post-tax income of the target user and the comprehensive benefit of the social e-commerce platform are considered. And the target tax calculation rule is recommended to the user and tax calculation is carried out after the confirmation information is obtained, so that the independent selection right of the user is ensured, and the tax calculation process is legal and compliant. According to the scheme, the accuracy, rationality and user satisfaction of CPS commission tax calculation are integrally improved, the optimal tax calculation scheme is automatically recommended for the user through the user portrait matching tax calculation rule, the selection cost of the user is reduced, and the overall tax compliance of the platform is also promoted.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a CPS commission settlement and tax calculation method, computer equipment, and storage medium. Background Art

[0002] With the rapid development of e-commerce, competition among e-commerce companies has become increasingly fierce. To improve their competitiveness, major e-commerce companies often use various methods to promote the products on their platforms. Among the many promotion methods, CPS (Cost Per Sales) pays promoters advertising fees based on actual transaction volume. This reduces promotion costs while ensuring the effectiveness of product promotion. Therefore, CPS has become one of the most popular promotion methods among major e-commerce companies. Users on social e-commerce platforms must file tax returns on the platform in addition to earning revenue. However, due to a lack of relevant knowledge, users often face the problem of excessively high costs when choosing a tax payment method. Summary of the Invention

[0003] The purpose of this application is to solve at least one of the above-mentioned technical defects, and in particular to provide a CPS commission settlement and tax calculation solution that reduces the user's selection cost.

[0004] First, this application provides a CPS commission settlement and tax calculation method, including:

[0005] Generate user profiles based on target users’ user information;

[0006] Match the user profile in the tax calculation rule library and filter out multiple tax calculation rules to be selected;

[0007] Evaluate the benefits of each proposed tax calculation rule;

[0008] Select the candidate tax calculation rule with the highest scheme benefit as the target tax calculation rule;

[0009] Recommend target tax calculation rules to target users;

[0010] In response to confirmation information of the target user regarding the target tax calculation rule, the CPS commission of the target user is taxed according to the target tax calculation rule.

[0011] In one embodiment, the user information includes user sub-information of multiple dimensions, and a user profile is generated based on the user information of the target user, including:

[0012] For any dimension of user sub-information, determine the label mapping relationship corresponding to the dimension;

[0013] According to the mapping relationship between user sub-information and tags, the user sub-information is mapped to the corresponding user tag to obtain the user profile.

[0014] In one embodiment, the tax calculation rule library includes multiple tax calculation rules and their corresponding applicable attributes. Matching is performed in the tax calculation rule library based on the user profile to screen out multiple tax calculation rules to be selected, including:

[0015] According to each user tag in the user portrait, match it with the applicable attributes corresponding to each tax calculation rule to obtain the corresponding matching degree of each tax calculation rule;

[0016] The rules with a matching degree greater than the matching degree threshold are determined as the tax calculation rules to be selected.

[0017] In one embodiment, the CPS commission settlement and tax calculation method further includes:

[0018] After evaluating the benefits of each tax calculation rule to be selected, the corresponding correlation coefficient is determined according to the benefits and matching degree of each tax calculation rule to be selected;

[0019] Determine the average correlation coefficient based on each correlation coefficient;

[0020] If the average correlation coefficient is lower than the preset correlation threshold, the matching threshold is lowered;

[0021] Otherwise, increase the matching threshold.

[0022] In one embodiment, evaluating the benefits of each tax calculation rule to be selected includes:

[0023] Determine the user groups corresponding to target users based on user portraits;

[0024] Determine the user benefits corresponding to each tax calculation rule based on the mapping relationship between the tax rate levels corresponding to the user groups and the tax rates in each tax calculation rule to be selected;

[0025] Determine the platform revenue corresponding to each candidate tax calculation rule based on the average operating cost corresponding to each candidate tax calculation rule;

[0026] Based on user revenue and platform revenue, evaluate the benefits of each proposed tax calculation rule.

[0027] In one embodiment, the CPS commission of the target user is taxed according to the target tax calculation rules, including:

[0028] Summarize the withdrawal data of target users during the tax calculation period;

[0029] Determine tax calculation data based on withdrawal data and target tax calculation rules;

[0030] Fill in the tax calculation data into the declaration form template to obtain the tax calculation declaration form;

[0031] Upload the tax return form.

[0032] In one embodiment, before uploading the tax return form, the following steps are also included:

[0033] Determine whether the total income recorded in the tax declaration form is consistent with the target user's total withdrawal amount, whether the withholding tax recorded in the tax declaration form is consistent with the target user's actual deduction amount, and whether the total self-invoicing amount recorded in the tax declaration form is consistent with the actual withdrawal amount;

[0034] If any inconsistency occurs, a manual screening prompt will be generated based on the inconsistent item, and the manual review process of the tax declaration form will be triggered based on the manual screening prompt.

[0035] In one embodiment, the CPS commission settlement and tax calculation method further includes:

[0036] Capture tax information announcements from the set data source according to the set crawling cycle;

[0037] Determine whether the tax calculation rule base needs to be updated based on tax information announcements;

[0038] If so, the tax calculation rule base will be updated according to the tax information announcement.

[0039] In one embodiment, the tax calculation rules in the tax calculation rule library include multiple preset rule fields, and whether the tax calculation rule library needs to be updated is determined based on the tax information announcement, including:

[0040] The tax information announcement and the preset rule field table are input into the target big model to instruct the target big model to determine whether the tax information announcement involves a change in the tax calculation rules. If so, the target big model will output the tax calculation rules to be matched. The preset rule field table includes all the preset rule fields that have appeared.

[0041] If the target large model outputs the tax calculation rules to be matched, the tax information announcement determines that the tax calculation rule library needs to be updated;

[0042] Otherwise, the tax calculation rule base does not need to be updated.

[0043] In one embodiment, updating the tax calculation rule base according to the tax information announcement includes:

[0044] Match the tax calculation rules to be matched with the tax calculation rules respectively;

[0045] If the tax calculation rule and the to-be-matched tax calculation rule have a first-proportion match in the preset rule fields and their corresponding values, a rule update prompt is generated based on the matched tax calculation rule and the to-be-matched tax calculation rule;

[0046] Otherwise, a new prompt is added based on the tax calculation rules to be matched;

[0047] Send rule update notifications or rule addition notifications to manual review;

[0048] The tax calculation rule library is updated based on the manual review results.

[0049] In one embodiment, the CPS commission settlement and tax calculation method further includes:

[0050] After the tax calculation rule library is updated, the updated tax calculation rules are backtested using historical tax calculation data;

[0051] If the backtest results do not meet the set requirements, a rule review prompt will be issued based on the updated tax calculation rules.

[0052] In a second aspect, the present application provides a CPS commission settlement and tax calculation method device, comprising:

[0053] User portrait module, used to generate user portraits based on the user information of target users;

[0054] The screening module is used to match the tax calculation rule library based on the user profile and filter out multiple tax calculation rules to be selected;

[0055] The benefit determination module is used to evaluate the benefits of each tax calculation rule to be selected;

[0056] A selection module is used to select the candidate tax calculation rule with the highest scheme benefit as the target tax calculation rule;

[0057] Recommendation module, used to recommend target tax calculation rules to target users;

[0058] The tax calculation module is used to calculate the tax on the CPS commission of the target user according to the target tax calculation rule in response to the confirmation information of the target user for the target tax calculation rule.

[0059] In a third aspect, the present application provides a computer device comprising one or more processors and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the steps of the CPS commission settlement and tax calculation method in any of the above embodiments of the claims are executed.

[0060] In a fourth aspect, the present application provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the CPS commission settlement and tax calculation method in any of the above embodiments.

[0061] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0062] Based on the CPS commission settlement and tax calculation method in this embodiment, by generating a user portrait based on the target user information, the user characteristics can be accurately portrayed, providing a data basis for the subsequent matching of tax calculation rules; based on the user portrait, the tax calculation rules to be selected are screened in the tax calculation rule library, realizing the accurate association between the tax calculation rules and the actual situation of the user. The benefits of each tax calculation rule to be selected are evaluated and the optimal rule is selected, taking into account the after-tax benefits of the target user and the comprehensive benefits of the social e-commerce platform. The target tax calculation rules are recommended to the user and the tax is calculated after the confirmation information is obtained, which guarantees the user's right to choose independently and ensures that the tax calculation process is legal and compliant. This solution improves the accuracy, rationality and user satisfaction of CPS commission tax calculation as a whole. By matching the tax calculation rules with user portraits, the optimal tax calculation plan is automatically recommended to the user, which reduces the user's selection cost and promotes the overall tax compliance of the platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0064] Figure 1 A flowchart of a CPS commission settlement and tax calculation method provided for one embodiment of the present application;

[0065] Figure 2 This is a flowchart of adjusting the matching threshold in one embodiment of the present application;

[0066] Figure 3 A schematic diagram of a process for evaluating the benefits of a scheme in one embodiment of the present application;

[0067] Figure 4 A schematic diagram of the process of generating a tax return form in one embodiment of the present application;

[0068] Figure 5 A schematic diagram of a process for updating a tax calculation rule library in one embodiment of the present application;

[0069] Figure 6 This is a flowchart of updating the tax calculation rule library in another embodiment of the present application;

[0070] Figure 7 This is a flowchart of updating the tax calculation rule library in another embodiment of the present application;

[0071] Figure 8 A diagram of the internal structure of a computer device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application and do not belong to all the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0073] This application provides a CPS commission settlement tax calculation method, please refer to Figure 1 , including steps S102 to S112.

[0074] S102: Generate a user profile based on the user information of the target user.

[0075] Target users are individuals or organizations that participate in the CPS model of social e-commerce platforms, promoting products through their social networks and earning commissions. User information encompasses multi-dimensional data, including but not limited to user registration information on the social e-commerce platform (such as age, gender, and region), historical shopping data (product types, purchase frequency, and average order value), social network data (number of friends, social activity, and social influence), and promotional behavior data (product types, promotion channels, and effectiveness data). User profiles are constructed based on user information through data mining and analysis techniques, abstracting and summarizing user characteristics and behavior patterns. In the social e-commerce sector, different target users, due to their varying attributes, have varying requirements and applicable rules for CPS commission settlement and tax calculation. By collecting and analyzing multi-dimensional information about target users and constructing user profiles, complex user characteristics can be quantified and categorized, providing an accurate basis for subsequent tax calculation rules. Serving as a bridge between user realities and tax calculation rules, user profiles can make the tax calculation process more tailored to user realities, improving accuracy and rationality.

[0076] S104: Match the user profile in the tax calculation rule library and filter out multiple tax calculation rules to be selected.

[0077] The tax calculation rule library is a database stored on the social e-commerce platform's servers, containing CPS commission tax calculation rules applicable to different user types. These rules cover various aspects, including tax rate calculation methods, tax incentive schemes, and tax base determination. The "selected tax calculation rules" are multiple rules potentially applicable to CPS commission tax calculation for target users, selected by matching user profiles with the tax calculation rule library. The numerous tax calculation rules stored in the tax calculation rule library are designed to meet the diverse needs of different users regarding CPS commission tax calculation. User profiles, as abstract representations of user characteristics, are matched against the rules in the tax calculation rule library to identify tax calculation rules that align with these user characteristics. This matching process compares user attributes, transaction behavior, and social characteristics contained in the user profiles with the applicable conditions set in the tax calculation rules. This matching and screening mechanism narrows the vast tax calculation rule library to a few rules relevant to the target user, providing specific criteria for subsequent program benefit evaluation and ensuring the targeted and effective selection of tax calculation rules.

[0078] Specifically, a rule-based matching approach can be used to convert user profile features into conditional expressions recognizable by the rule engine, and then match them with the applicable conditions of each rule in the tax calculation rule library. For example, using the Drools rule engine, conditions such as transaction amount and transaction frequency in the user profile can be matched with the tax calculation rules for different transaction sizes in the tax calculation rule library.

[0079] S106, evaluating the benefits of each tax calculation rule to be selected.

[0080] It can be understood that solution benefits refer to the benefits derived from different taxation methods for target users when applying a proposed tax calculation rule to their CPS commissions, as well as the comprehensive benefits of the proposed tax calculation rule on the social e-commerce platform in terms of cost and convenience. The evaluation process requires comprehensive consideration of multiple factors, including but not limited to the amount of tax payable, the amount of tax exemptions and reductions from tax incentives, and the impact of tax calculation complexity. Different proposed tax calculation rules, due to their varying tax rates, tax calculation methods, and preferential schemes, will result in varying after-tax benefits for target users and the overall benefits of the platform. By evaluating the solution benefits of various proposed tax calculation rules, we can compare the pros and cons of different rules from multiple perspectives, providing a quantitative basis for selecting the optimal tax calculation rule. For target users, solution benefits are directly related to their actual commission income. For social e-commerce platforms, a reasonable tax calculation rule must not only protect user interests but also maximize platform revenue while maintaining tax compliance, while also improving user satisfaction and retention. For example, a potential tax calculation rule might offer a lower tax rate but a complex calculation method, potentially increasing user operational costs and risking misunderstandings, thus reducing user experience. Meanwhile, another rule might offer a slightly higher tax rate but be simpler to calculate, making it easier for users to understand and accept. Therefore, a comprehensive assessment of these factors is necessary to accurately determine the benefits of each potential tax calculation rule.

[0081] S108, selecting the candidate tax calculation rule with the highest scheme benefit as the target tax calculation rule.

[0082] It can be understood that the target tax calculation rule is the rule that is most suitable for the CPS commission tax calculation of target users, which is screened out from multiple tax calculation rules after evaluating the benefits of each tax calculation rule. This rule can bring the best comprehensive benefits to target users and social e-commerce platforms under the premise of ensuring tax compliance.

[0083] S110, recommending target tax calculation rules to target users.

[0084] Recommendation, as it can be understood, refers to a social e-commerce platform presenting a selected target tax calculation rule to target users through specific messaging methods, enabling them to understand the specific content, advantages, and impact of the rule on the tax calculation of their CPS commissions. The recommendation process should consider user acceptance and information acquisition habits, utilizing appropriate channels and methods for information delivery. Social e-commerce platforms can recommend the target tax calculation rule to target users through in-site messaging, email, app push notifications, and other methods. The message content should detail the specific terms, calculation method, and expected after-tax benefits of the rule. The purpose of recommending the target tax calculation rule to target users is to help them understand and accept the rule, thereby increasing their satisfaction and cooperation with the tax calculation process. In the social e-commerce environment, users are particularly concerned about the tax calculation method of their CPS commissions. By recommending the target tax calculation rule, users can clearly understand how their commissions will be taxed and the actual benefits they can achieve by adopting the rule.

[0085] S112 , in response to the target user's confirmation information regarding the target tax calculation rule, the target user's CPS commission is taxed according to the target tax calculation rule.

[0086] Confirmation information refers to the target user's feedback to the social e-commerce platform through interactive means provided by the platform (such as clicking a confirmation button or submitting written confirmation documents) after receiving the target tax calculation rules recommended by the platform, indicating their approval and agreement to the use of these rules for CPS commission tax calculation. Tax calculation refers to the process of calculating the taxable amount and tax treatment of commissions earned by the target user through the CPS model over a certain period of time, in accordance with the tax rate, tax calculation method, and preferential scheme stipulated in the target tax calculation rules.

[0087] Based on the CPS commission settlement and tax calculation method in this embodiment, by generating a user portrait based on the target user information, the user characteristics can be accurately portrayed, providing a data basis for the subsequent matching of tax calculation rules; based on the user portrait, the tax calculation rules to be selected are screened in the tax calculation rule library, realizing the accurate association between the tax calculation rules and the actual situation of the user. The benefits of each tax calculation rule to be selected are evaluated and the optimal rule is selected, taking into account the after-tax benefits of the target user and the comprehensive benefits of the social e-commerce platform. The target tax calculation rules are recommended to the user and the tax is calculated after the confirmation information is obtained, which guarantees the user's right to choose independently and ensures that the tax calculation process is legal and compliant. This solution improves the accuracy, rationality and user satisfaction of CPS commission tax calculation as a whole. By matching the tax calculation rules with user portraits, the optimal tax calculation plan is automatically recommended to the user, which reduces the user's selection cost and promotes the overall tax compliance of the platform.

[0088] In one embodiment, user information includes user sub-information across multiple dimensions. Generating a user profile based on the target user's user information includes: determining, for each dimension of user sub-information, a tag mapping relationship corresponding to that dimension. Based on the user sub-information and tag mapping relationship, mapping the user sub-information to the corresponding user tag to generate a user profile.

[0089] It can be understood that user sub-information is the specific data representation of user information along specific dimensions. It is the fundamental unit of a multi-dimensional user information system, encompassing specific data across dimensions such as basic attributes, transaction behavior, and social attributes, such as age, transaction amount, and number of followers. Dimensions serve as criteria for categorizing and classifying user information, deconstructing user information from different perspectives for targeted analysis and processing. Label mapping relationships are pre-defined sets of logically defined rules that convert user sub-information into standardized user labels according to specific criteria. They are the key link in transforming raw data into structured labels. This mapping relationship is based on an understanding of social e-commerce business requirements and tax treatment rules. By establishing a link between data and labels, user information is abstracted and refined. In the context of social e-commerce CPS commission settlement and tax calculation, the diversity and complexity of user information makes it difficult to directly use raw data to match tax calculation rules. By dividing user information into sub-information along different dimensions and defining corresponding label mapping relationships for each dimension, user information can be organized and standardized. For example, within the historical income dimension, label mappings can convert users with different income levels into labels such as "high-income user" and "middle-income user." These independent yet collaborative label mappings across dimensions capture a comprehensive user profile, providing structured data support for precise matching of tax calculation rules. This allows subsequent tax calculations to better align with users' actual circumstances, improving both accuracy and rationality.

[0090] Social e-commerce platforms can use relational database management systems (such as MySQL and Oracle) to store and manage user sub-information and tag mappings. In database design, create a separate table for each dimension to store user sub-information and the corresponding tag mappings. For example, create a "transaction behavior table" containing fields such as "transaction amount" and "transaction frequency" to store user sub-information, while also setting up "amount tag mapping" and "frequency tag mapping" fields to record the corresponding tag mappings. Using database query statements (such as SQL SELECT and JOIN statements), you can quickly retrieve and obtain user sub-information for a specific dimension and its corresponding tag mappings.

[0091] User tags are standardized representations of user sub-information, resulting from the distillation and summary of user sub-information. They concisely and clearly reflect a user's core characteristics or attributes along a specific dimension and are a fundamental component of user profiles. User profiles, by integrating user tags across multiple dimensions, are constructed to comprehensively and accurately depict a user's comprehensive characteristics, needs, and behavioral tendencies in social e-commerce scenarios. This highly abstract and summarized representation of complex user information provides intuitive and effective data support for social e-commerce platforms in CPS commission settlement, tax calculation, and other business decision-making. Driven by rules based on tag mapping relationships, user sub-information is transformed according to established standards, transforming the previously scattered and disorganized raw data into a set of tags with clear business meaning and classification. These tags describe users from different perspectives. For example, basic attribute tags reflect user identity, transaction behavior tags reflect user business activity patterns, and social attribute tags demonstrate user social influence. Once all dimensions of user sub-information are mapped, these tags are integrated to form a complete user profile. As an integrated whole, the user profile comprehensively captures user characteristics, providing a clear and accurate basis for selecting appropriate tax calculation rules from the tax calculation rule library.

[0092] In one embodiment, a tax calculation rule library includes multiple tax calculation rules and their corresponding applicable attributes. A matching process is performed in the tax calculation rule library based on a user profile to screen multiple tax calculation rules for selection. This process includes matching user tags in the user profile with the applicable attributes corresponding to each tax calculation rule to obtain a matching degree for each tax calculation rule. Rules with matching degrees exceeding a matching degree threshold are determined as candidate tax calculation rules.

[0093] The tax calculation rule library is a database that stores all tax calculation rules on a social e-commerce platform. Each tax calculation rule corresponds to a set of applicable attributes, which quantify or categorize the applicable conditions of the rule and are used to define the scope of the rule. The applicable attributes correspond semantically and structurally to user tags, facilitating matching. The matching degree is a quantitative indicator that measures the degree of fit between a user profile and a tax calculation rule. Calculated using a specific algorithm, it reflects the extent to which a user meets the applicable conditions of a particular tax calculation rule. In the social e-commerce CPS commission settlement and tax calculation scenario, the tags in the user profile characterize the user from multiple dimensions, while the applicable attributes of the tax calculation rule define the conditions under which the rule takes effect. Matching the two allows filtering out rules relevant to user characteristics from a vast pool of tax calculation rules. The matching results of each user tag and the applicable attributes constitute a component of the matching degree of the tax calculation rule. The final matching degree is determined by combining the matching results of all tags. This mechanism makes the tax calculation rule screening process more precise and systematic, avoiding the subjectivity and inefficiency of manual screening and providing a reliable basis for the subsequent selection of the optimal tax calculation rule. A rule-based matching algorithm can be used to match labels with applicable attributes. For example, for numeric labels and attributes, a similarity score is calculated; for categorical labels and attributes, whether they are an exact match or belong to the same category is determined. The weighted sum of the matching scores for each label is used to determine the overall matching degree for the tax calculation rule.

[0094] The matching threshold is a pre-set critical value used to determine whether the user profile and the tax calculation rules have reached an acceptable level of match. The candidate tax calculation rules are the set of tax calculation rules that, after matching screening, are deemed highly relevant to the user's characteristics. These rules will enter the subsequent evaluation and selection phase. The purpose of setting a matching threshold is to control the number of candidate rules while ensuring the relevance of the tax calculation rules to the user profile, preventing an excessive number of irrelevant rules from entering the subsequent process and improving system efficiency. Threshold screening allows us to focus on the tax calculation rules that best match the user's characteristics, reducing the complexity of subsequent evaluation.

[0095] In one embodiment, the CPS commission settlement tax calculation method is as follows: Figure 2 , also including steps S202 to S208.

[0096] S202: After evaluating the program benefits of each candidate tax calculation rule, determine the corresponding correlation coefficient according to the program benefits and matching degree of each candidate tax calculation rule.

[0097] It can be understood that the correlation coefficient is a quantitative value used to measure the degree of association between the solution benefits and the matching degree of the proposed tax calculation rules. It reflects the synergistic relationship between the two. The higher the coefficient, the closer the correlation between the solution benefits and the matching degree, which means that the tax calculation rules can bring better overall benefits while fitting the user's characteristics. In the CPS commission settlement and tax calculation process, solution benefits and matching degree are two important dimensions for evaluating proposed tax calculation rules. Solution benefits reflect the value of the rules at the economic and business levels, while matching degree reflects the degree of adaptation of the rules to the user's actual situation. Determining the correlation coefficient aims to reveal the inherent connection between these two dimensions, thereby more comprehensively evaluating the proposed tax calculation rules. For example, if a proposed tax calculation rule has a high matching degree but a low solution benefit, it means that although the rule meets the user's characteristics, it cannot bring the ideal benefits to the user and the platform in actual tax calculation. Conversely, if the solution benefit is high but the matching degree is low, there may be a risk of unstable rule applicability. By calculating the correlation coefficient, these two key factors can be combined for analysis, providing a more accurate basis for the subsequent screening and optimization of tax calculation rules, ensuring that the final selected tax calculation rules can both meet user needs and maximize the interests of the platform and users.

[0098] S204: Determine an average correlation coefficient based on the correlation coefficients.

[0099] The average correlation coefficient is the statistical average of the correlation coefficients of all candidate tax calculation rules. It represents the overall correlation between solution benefits and compatibility across the entire set of candidate tax calculation rules. This value measures the overall performance of the candidate tax calculation rules in balancing user adaptability and overall benefits, and serves as an important reference for subsequent decision-making. While the correlation coefficient of each candidate tax calculation rule only reflects the correlation between the solution benefits and compatibility of that rule itself, the average correlation coefficient provides a comprehensive, macro-level assessment of the overall quality of all candidate tax calculation rules. By calculating the average correlation coefficient, we can understand the overall effectiveness of the currently selected candidate tax calculation rules in balancing user adaptability and overall benefits. A high average correlation coefficient indicates that the candidate tax calculation rules generally strike a good balance between user characteristics and actual benefits. Conversely, a low average correlation coefficient indicates that some of the candidate tax calculation rules may not strike an ideal balance between these two factors, necessitating adjustments to the screening strategy. As a quantitative overall indicator, the average correlation coefficient provides an objective basis for the subsequent adjustment of the matching threshold according to actual conditions, which helps to optimize the screening process of tax calculation rules and improve the overall quality of the tax calculation plan.

[0100] S206: If the average correlation coefficient is lower than the preset correlation threshold, lower the matching threshold.

[0101] S208: Otherwise, increase the matching threshold.

[0102] It can be understood that the preset relevance threshold is a reference value set by social e-commerce platforms based on business needs and historical experience. It is used to determine whether the overall quality of the current set of candidate tax calculation rules meets expected standards. The matching threshold is a critical value set when screening candidate tax calculation rules. A rule is selected as a candidate tax calculation rule only when its matching degree with the user profile exceeds this threshold. Lowering the matching threshold relaxes the screening criteria for tax calculation rules, allowing more rules that were originally not selected due to insufficient matching to enter the selection range. The average relevance coefficient reflects the overall performance of the candidate tax calculation rules in balancing user adaptability and overall benefits. When the average relevance coefficient is lower than the preset relevance threshold, it indicates that the currently selected candidate tax calculation rules do not effectively balance user characteristics and actual benefits. This may be due to the matching threshold being set too high, resulting in the exclusion of some tax calculation rules with lower matching degrees but higher returns. By lowering the matching threshold, the range of candidate tax calculation rules can be expanded, introducing more potential tax calculation rules for subsequent evaluation, and increasing the possibility of finding more optimal tax calculation rules. This allows us to more fully explore tax calculation rules with higher plan benefits, optimize tax calculation plans, improve users' after-tax returns and the platform's overall efficiency, and make the tax calculation rule screening process more consistent with actual business needs, while ensuring a certain level of user compatibility. Specifically, when the average correlation coefficient falls below a preset correlation threshold, the system automatically lowers or raises the matching threshold by a fixed step size, such as 5%.

[0103] In one embodiment, the benefits of each tax calculation rule are evaluated. Figure 3 , including steps S302 to S308.

[0104] S302: Determine the user group corresponding to the target user based on the user portrait.

[0105] User grouping is a classification method that groups users with common characteristics or behavior patterns into the same set based on the similarity of their user profiles. Each user group corresponds to specific business attributes and tax treatment rules, serving as a critical intermediary between individual user characteristics and the tax calculation rule system. In the context of social e-commerce CPS commission settlement and tax calculation, user groups exhibit diverse characteristics, and factors such as transaction volume and social influence influence the applicable tax calculation rules. Determining user groups based on user profiles allows for a complex and diverse user population to be categorized based on similar characteristics, allowing users with the same or similar characteristics to apply a relatively unified tax calculation strategy. User grouping, as the fundamental unit for tax calculation rule screening and application, greatly improves the efficiency and accuracy of tax calculation rule matching. Social e-commerce platforms often utilize clustering algorithms to classify user profiles. For example, the K-means clustering algorithm sets a specific number of clusters (i.e., the number of user groups) and calculates the distance between feature vectors in the user profiles to group similar users together.

[0106] S304: Determine the user income corresponding to each tax calculation rule to be selected based on the tax rate level mapping relationship corresponding to the user group and the tax rate in each tax calculation rule to be selected.

[0107] It can be understood that the tax rate level mapping relationship is the correspondence between tax rates and tax rate levels pre-set by social e-commerce platforms based on the tax rate distribution within the same user group. The tax rate in each candidate tax calculation rule refers to the ratio specified by each rule for calculating the tax payable, among the multiple candidate tax calculation rules selected. User revenue refers to the tax rate of the target user within the same user group after the tax payable is calculated using a candidate tax calculation rule. The lower the tax rate, the higher the user revenue of the user among similar users. For example, the tax rate level mapping relationship corresponding to a user group is set as follows: tax rates >10% are high, 5%-10% are medium, and <5% are low. If a candidate tax calculation rule sets a tax rate of 4% for this group of users, then based on the mapping relationship, it is determined that they belong to the low tax rate level, and the user revenue under this rule is relatively high. This judgment mechanism uses a unified standard to convert abstract tax rate values ​​into intuitive profit evaluations, enabling the platform to quickly compare the impact of different candidate tax calculation rules on user profits, laying the foundation for subsequent screening of rules that can both ensure that users obtain relatively preferential tax calculation conditions within their group and comply with the platform's overall tax calculation strategy, thereby achieving coordinated optimization of user profits and platform tax calculation management.

[0108] S306: Determine the platform revenue corresponding to each of the tax calculation rules to be selected based on the average operating cost corresponding to each of the tax calculation rules to be selected.

[0109] It can be understood that average operating costs refer to the average expenses incurred by social e-commerce platforms for providing related services and support when implementing specific tax calculation rules, including system maintenance costs, customer service costs, tax compliance costs, etc. Platform revenue in this context does not represent actual economic income, but rather a quantitative indicator used to reflect the impact of different proposed tax calculation rules on various aspects of platform operations. This indicator assesses the relative importance and impact of each rule on the platform by comprehensively considering the costs and resources involved in implementing the rule, as well as its potential impact on the platform's business processes and user experience.

[0110] Different tax calculation rules may correspond to different operating costs. Average operating cost, a key factor in measuring the resource consumption of rule execution, is an important basis for evaluating platform revenue (i.e., the impact of the rule). For example, a potential tax calculation rule that requires complex algorithmic support and frequent manual review has a high average operating cost. When evaluating platform revenue, this means that the rule occupies and potentially puts greater pressure on platform operating resources, resulting in a greater relative impact. Conversely, rules with low operating costs have a relatively smaller impact on the platform. By correlating average operating cost with platform revenue, it is possible to systematically analyze the status and role of each potential tax calculation rule in the platform's operating system, providing key data for subsequent comprehensive evaluation of tax calculation rule solutions. This helps the platform determine which rules are more conducive to maintaining efficient and stable operations and which rules may impose greater management and resource burdens, thereby optimizing the tax calculation rule selection strategy and achieving efficient platform operations and cost rationalization.

[0111] S308: Evaluate the benefits of each tax calculation rule to be selected based on user benefits and platform benefits.

[0112] It can be understood that in this embodiment, user benefits are a quantitative indicator reflecting the impact of different proposed tax calculation rules on the target user's tax benefits and relative benefits within their group; platform benefits are a quantitative indicator reflecting the impact of different proposed tax calculation rules on various aspects of platform operations. Solution benefits are a quantitative result that combines user benefits and platform benefits, used to comprehensively evaluate the overall merits of each proposed tax calculation rule. It is not actual economic benefits, but rather reflects the combined impact of each rule in balancing user needs and platform operations. In the CPS commission settlement and tax calculation method, user benefits and platform benefits reflect the characteristics and impact of the proposed tax calculation rules from different dimensions. User benefits focus on the impact of the rules on the user's tax benefits within their group, while platform benefits focus on the impact of the rules on platform operations. Combining these two factors to evaluate solution benefits allows for a more comprehensive assessment of the combined impact of each proposed tax calculation rule on users and the platform. For example, a proposed tax calculation rule may have high user benefits, making it more attractive to users, but low platform benefits, meaning that implementing it would place significant operational pressure on the platform. Another rule, however, may strike a better balance between the two. By comprehensively evaluating the benefits of the plan, we can avoid the limitations of single-dimensional evaluation and screen out tax calculation rules that both meet user needs and are in line with the actual platform operations.

[0113] In one embodiment, the CPS commission of the target user is taxed according to the target tax calculation rules, see Figure 4 , including steps S402 to S408.

[0114] S402, summarizing the withdrawal data of the target user in the tax calculation period.

[0115] As can be understood, a tax calculation period is a pre-defined time period used for tax statistics and calculation, such as a month or a quarter. It provides a time dimension for tax calculation. Withdrawal data refers to the records of target users withdrawing commissions earned through the CPS model from their social e-commerce platform accounts to external accounts (such as bank accounts or third-party payment accounts) during the tax calculation period. This data includes information such as withdrawal amount, withdrawal time, and withdrawal channel. This data is a crucial basis for tax calculation. In the social e-commerce CPS commission settlement and tax calculation system, accurate tax calculation relies on complete and accurate business data. Aggregating target users' withdrawal data during the tax calculation period provides a comprehensive overview of their actual commission income during that time period. Since users' commission income is not realized all at once but rather spread across multiple withdrawals within the tax calculation period, aggregating withdrawal data can consolidate scattered transaction information into comprehensive tax calculation data. For example, if a user withdraws multiple amounts in a month, only by aggregating these withdrawal data can we accurately determine the user's total commission income for that month, providing accurate data support for the subsequent calculation of tax payable based on the targeted tax calculation rules. This step is the starting point of the tax calculation process. The integrity and accuracy of the data in this step directly affects the results of subsequent tax calculations. It is a key prerequisite for ensuring the smooth progress of tax calculations and the authenticity and reliability of tax calculation results.

[0116] S404: Determine tax calculation data based on the withdrawal data and target tax calculation rules.

[0117] Tax calculation data refers to the key data used to determine tax payable, including the tax base, applicable tax rate, and taxable income, derived through calculation and processing based on the target user's withdrawal data and the selected target tax calculation rules. The target tax calculation rules are the most appropriate rules for calculating CPS commissions for the target user, selected after a series of screening and evaluation processes. They include specific details such as the tax rate calculation method, tax incentive scheme, and tax basis. Withdrawal data reflects the target user's actual commission income, while the target tax calculation rules specify how these income will be taxed. Determining tax data based on withdrawal data and target tax calculation rules involves matching actual business data with the tax calculation rules and calculating them. Specifically, the tax base is first determined based on the target tax calculation rules, for example, using the withdrawal amount as the tax base or determining the tax base after deducting certain costs and expenses. Then, tax data such as taxable income is calculated based on the tax base and the applicable tax rate specified in the rules. For example, if the target tax calculation rules stipulate that the portion of a user's withdrawal amount exceeding a certain threshold is taxed at a specific rate, then it is necessary to determine whether the withdrawal amount exceeds the threshold based on the withdrawal data and calculate the corresponding tax data. This step is the core of the tax calculation process. By combining specific business data with abstract tax calculation rules, it achieves the conversion from user revenue data to key tax calculation data, providing a direct basis for accurately calculating the tax payable and ensuring the compliance and accuracy of the tax calculation process.

[0118] S406: Fill in the tax calculation data into the declaration form template to obtain the tax calculation declaration form.

[0119] A tax return form template is a pre-designed electronic document or form based on tax regulations or industry standards. It contains a fixed format and fields for entering tax calculation information, such as taxpayer basic information, tax calculation period, tax calculation data, and tax payable. The tax calculation return form is a complete tax filing document generated by accurately entering the tax calculation data into the corresponding fields of the return form template. It serves as the official tax filing document for submitting tax returns to the relevant authorities, reflecting the target user's CPS commission tax calculation status during the tax calculation period. After calculating the tax calculation data, accurately entering it into the return form template to create the tax calculation return form is done to present the tax calculation results in a standardized and unified format to meet the reporting requirements of relevant authorities. The design of the return form adheres to tax administration regulations and standards, ensuring the completeness and accuracy of tax calculation information. By filling in the tax calculation data according to the template requirements, relevant authorities can quickly and clearly access the taxpayer's tax calculation information, facilitating tax audits and management. Social e-commerce platforms can use office automation software (such as Microsoft Excel or WPS Spreadsheets) to design declaration form templates and programming languages ​​(such as VBA or Python's openpyxl library) to automatically fill in tax calculation data. After the tax calculation is completed, the program automatically reads the tax data and fills it into the corresponding cells of the declaration form template.

[0120] S408, upload the tax declaration form.

[0121] Understandably, uploading the tax return form is the final, critical step in the entire CPS commission settlement and tax calculation process. Its purpose is to promptly and accurately transmit taxpayers' tax declaration information to the tax authorities, fulfilling tax filing obligations. In the context of modern e-government and information-based tax management, uploading tax return forms online has digitized and automated the tax filing process.

[0122] In one embodiment, before uploading the tax return form, the following steps are performed: 1. Check whether the total income recorded in the tax return form is consistent with the target user's total withdrawal amount; 2. Check whether the withheld taxes recorded in the tax return form are consistent with the target user's actual deduction amount; 3. Check whether the total self-invoiced invoices recorded in the tax return form are consistent with the actual withdrawal amount. If any of these items are inconsistent, a manual screening prompt is generated based on the inconsistent item, which triggers the manual review process of the tax return form.

[0123] As you can understand, the total withdrawal amount is the sum of all commission withdrawals made by the target user through the social e-commerce platform during the tax calculation period. This data is derived from the platform's withdrawal records. Withholding tax is the amount of tax payable that the social e-commerce platform pre-deducts in accordance with the target tax calculation rules during the commission withdrawal process. The actual deduction amount is the total amount of tax deducted from the withdrawal amount by the platform on each withdrawal. The total self-invoiced amount is the total amount of invoices issued by the target user in accordance with tax regulations during the tax calculation period. The actual withdrawal amount is the commission amount that the user ultimately successfully withdraws to an external account. In the CPS commission settlement and tax calculation process, the tax return form is the core document for tax filings, and the accuracy of its data is crucial. By comparing total revenue with total withdrawal amount, withholding tax with actual deduction amount, and self-invoiced amount with actual withdrawal amount, we can fully verify that the data in the tax return form is consistent with actual business transactions. Comparing total revenue with total withdrawals verifies the accuracy of the tax base; comparing withheld taxes with actual deductions ensures the accuracy of tax calculations and deductions; and comparing self-invoiced totals with actual withdrawals helps verify the matching of invoices with actual transactions. Inconsistencies in these key data points may indicate errors in the tax calculation process, such as miscalculation, improper application of rules, or data transmission errors. Promptly identifying and addressing these inconsistencies can avoid erroneous tax declarations, ensure the accuracy and compliance of tax calculations, and mitigate tax risks.

[0124] Manual screening prompts are automatically generated by the system when discrepancies are detected between tax return data and actual business data. These prompts contain detailed information about the discrepancies and guide manual reviewers to quickly identify the issue. For example, a prompt might read, "The total income and total withdrawal amount are inconsistent, with a difference of XX yuan. Please verify the application of tax calculation rules and the data calculation process." The manual review process involves professional tax personnel or platform finance staff manually checking and verifying the tax return and related business data after receiving a manual screening prompt. This process aims to identify the root cause of the data discrepancy through manual intervention and make corrections to ensure the accuracy of the tax return. When the system detects data discrepancies in the tax return, it automatically generates a manual screening prompt and triggers the manual review process. This is to introduce a manual intervention mechanism based on automated data processing to overcome the limitations of the system. Because the causes of data inconsistencies are complex, involving multiple factors such as rule understanding, special business processing, and data anomalies, relying solely on the system cannot fully and accurately identify and resolve the issues. Manual screening prompts clearly and intuitively present issues to professionals, allowing them to quickly understand the key issues. The manual review process, leveraging professionals' business knowledge and experience, conducts in-depth analysis and investigation of tax return forms and related data, accurately identifying the causes of data inconsistencies—such as errors in the application of tax calculation rules, data entry errors, and system calculation vulnerabilities—and implementing targeted corrections. This combination of automated detection and manual intervention effectively improves the accuracy of tax calculation data, ensures the reliability of tax returns, and reduces tax risks and corporate losses caused by data errors.

[0125] In one embodiment, the CPS commission settlement tax calculation method is as follows: Figure 5 , also including steps S502 to S506.

[0126] S502: Capture tax information announcements from a set data source according to a set capture cycle.

[0127] Understandably, a set crawling cycle is a pre-configured interval for social e-commerce platforms to automatically obtain tax information announcements, such as daily or weekly, to balance the timeliness of information acquisition with system resource consumption. The set data source refers to the platform or organization that publishes the tax information announcements. These data sources are the original source of the relevant tax information. Tax information announcements are official documents issued by relevant departments regarding tax adjustments, changes in tax calculation rules, and tax filing requirements. They serve as the basis for social e-commerce platforms to update their tax calculation rule bases.

[0128] By regularly crawling tax information announcements from authoritative data sources, social e-commerce platforms can stay up-to-date on the latest tax developments. This is essential for ensuring that the platform's tax calculation rule base remains synchronized with current tax rules. For example, if the tax authorities issue a new tax rate adjustment announcement, the platform may still calculate taxes according to the old rules if it fails to obtain this information in a timely manner. This can lead to tax errors and create tax risks for the platform and its users. Setting a reasonable crawling cycle ensures information timeliness while avoiding the waste of system resources caused by overly frequent crawling operations, thus achieving a balance between information acquisition efficiency and system operating costs.

[0129] S504: Determine whether the tax calculation rule library needs to be updated based on the tax information announcement.

[0130] It can be understood that determining whether the tax calculation rule base needs to be updated means analyzing the content of the tax information announcement to determine whether it contains content that affects the existing tax calculation rules, and thus deciding whether the tax calculation rule base needs to be modified and improved. The tax information announcement contains a wide range of content, and not all announcements are directly related to the social e-commerce CPS commission tax calculation rules. By conducting targeted analysis of the announcement content, it is possible to accurately identify which announcements involve changes to the tax calculation rules and which are not related to the platform's business. For example, the announcement may involve corporate income tax adjustments, but the CPS commission tax calculation of the social e-commerce platform mainly involves other taxes such as personal income tax. At this time, it is necessary to determine whether the announcement has an impact on the platform's current tax calculation rules. Only when the content of the announcement does involve changes related to the tax calculation rules will the update process of the tax calculation rule base be triggered, avoiding unnecessary system operations and improving the efficiency and accuracy of the platform's tax calculation management.

[0131] S506: If yes, update the tax calculation rule library according to the tax information announcement.

[0132] It can be understood that updating the tax calculation rule library means that when it is determined that the tax calculation rule library needs to be updated according to the tax information announcement, the relevant content in the tax calculation rule library is modified, added or deleted according to the content of the announcement to ensure that the tax calculation rule library is consistent with the latest tax rules.

[0133] In one embodiment, the tax calculation rules in the tax calculation rule library include multiple preset rule fields. It is determined whether the tax calculation rule library needs to be updated based on the tax information announcement. Figure 6 , including steps S602 to S606.

[0134] S602: Input the tax information announcement and the preset rule field table into the target macro model, instructing the target macro model to determine whether the tax information announcement involves a change in tax calculation rules and, if so, output the tax calculation rules to be matched. The preset rule field table includes all occurrences of the preset rule fields.

[0135] As you can understand, the target large-scale model is a large-scale language model trained using deep learning technology. It possesses powerful natural language understanding and reasoning capabilities, capable of performing semantic analysis and logical judgment on input text. In this process, it is used to identify changes to tax calculation rules in tax information announcements. The preset rule field table is a collection of all historical tax calculation rule-related fields compiled by social e-commerce platforms, such as tax rates, tax thresholds, deduction standards, and tax calculation methods. This helps the large-scale model more accurately identify rule changes in announcements. To-be-matched tax calculation rules are new or adjusted rules extracted from the announcement content when the target large-scale model determines that a tax information announcement involves a change in tax calculation rules. These rules need to be matched against the existing tax calculation rule library. Traditional tax information analysis methods based on keyword matching have limitations and struggle to understand the deep semantics and complex logical relationships within announcement text. However, through training on a large corpus, the target large-scale model is able to capture semantic connections and contextual information within the text, enabling it to more accurately determine whether a tax information announcement truly involves a change in tax calculation rules. The preset rule field table provides the large-scale model with domain-specific prior knowledge, helping it focus on content related to tax calculation rules and improving recognition accuracy.

[0136] Social e-commerce platforms can use pre-trained large-scale language models (such as GPT-3 and BERT) as the target large-scale model and fine-tune it for the domain. First, they collect a large amount of tax-related text data, including historical tax information announcements, tax calculation rule documents, and tax regulations, to build a domain-specific corpus. This corpus is then used to fine-tune the pre-trained large-scale model to better understand tax terminology and semantics. In practice, the target large-scale model is called via an API using the tax information announcement text and a table of pre-set rule fields as input. The large-scale model performs word segmentation, encoding, and semantic analysis on the input text to determine whether the announcement involves a change in tax calculation rules. If so, the large-scale model uses entity recognition and relationship extraction techniques to extract key rule elements from the announcement, such as tax type, tax rate, and applicable conditions, and then generates the output tax calculation rules to match.

[0137] S604: If the target large model outputs the tax calculation rules to be matched, the tax information announcement determines that the tax calculation rule library needs to be updated.

[0138] It's understandable that if the target large model doesn't output unmatched tax calculation rules, it means the announcement is unrelated to the platform's tax calculation rules. However, if the target large model outputs unmatched tax calculation rules, it indicates that it has identified changes to the tax calculation rules from the tax information announcement. These changes may involve tax rate adjustments, changes in tax calculation methods, and updates to tax incentive programs, directly impacting the tax calculation basis for CPS commission settlements on social e-commerce platforms. To ensure that the platform's tax calculation rule base is consistent with the latest tax regulations and avoid tax risks caused by the use of outdated rules, the update process for the tax calculation rule base should be triggered when the large model outputs unmatched tax calculation rules.

[0139] S606: Otherwise, it is determined that the tax calculation rule library does not need to be updated.

[0140] It's understandable that if the target large model doesn't output a matching tax calculation rule, it means that, based on its analysis of the tax information announcement, it has determined that the announcement does not involve a change to the tax calculation rules. In this case, maintaining the existing tax calculation rule base without updating it can avoid unnecessary system overhead and potential update risks. For example, if the announcement only involves optimizing the tax filing process or improving tax services without affecting the tax calculation rules themselves, there's no need to update the tax calculation rule base. This intelligent decision-making mechanism improves system efficiency and ensures the stability and reliability of the tax calculation rule base.

[0141] In one embodiment, the tax calculation rule base is updated according to the tax information announcement, see Figure 7 , including steps S702 to S710.

[0142] S702, matching the tax calculation rules to be matched with the tax calculation rules respectively.

[0143] This step compares the consistency or similarity between the to-be-matched tax calculation rule and existing tax calculation rules in terms of pre-set rule fields and their corresponding values. This determines the relationship between the two and whether the to-be-matched tax calculation rule is an update to an existing rule or a completely new rule. When updating the tax calculation rule library based on tax information announcements, matching the to-be-matched tax calculation rule with existing rules is fundamental for accurately identifying the type and scope of rule changes. Since the to-be-matched tax calculation rule extracted from the tax information announcement may be a partial adjustment to an existing rule or a completely new rule, the matching operation clearly defines its relationship to the existing rule. If the to-be-matched tax calculation rule shows a high degree of similarity with an existing rule, it may be an update to that rule. If the similarity is low or even non-existent, it indicates that a new rule is needed. This matching mechanism enables the platform to handle rule adjustments resulting from tax changes in an orderly manner, avoiding blind updates or missed important rule changes. Rule matching can be performed using algorithms based on string matching and semantic similarity. Python text processing libraries (such as NLTK and spaCy) are used to perform word segmentation, part-of-speech tagging, and named entity recognition on the rule text, extracting the pre-set rule fields. Edit distance algorithms (such as the Levenshtein distance) are used to calculate the degree of difference between the field text of the tax calculation rule to be matched and the tax calculation rule, which serves as a measure of matching. Furthermore, word embedding models (such as Word2Vec and BERT) are introduced to map the rule fields into vector representations. Semantic similarity is assessed by calculating the cosine similarity between these vectors.

[0144] S704: If the tax calculation rule and the tax calculation rule to be matched have a first ratio of preset rule fields and their corresponding values ​​matched, a rule update prompt is generated according to the matched tax calculation rule and the tax calculation rule to be matched.

[0145] It can be understood that the first ratio is a threshold ratio, such as 60% or 70%, pre-set by the social e-commerce platform to determine whether a tax calculation rule and the to-be-matched tax calculation rule are in an update relationship. When the number of matching preset rule fields and their corresponding values ​​reaches or exceeds this ratio, the to-be-matched tax calculation rule is considered an update to the existing tax calculation rule. A rule update prompt is a system-generated message containing detailed rule change information based on the matching tax calculation rule and the to-be-matched tax calculation rule. It informs manual reviewers that an update to the existing tax calculation rule is required, including the specific update content and direction. For example, the original rule "Individual Commission Income Tax Rate 10%" should be updated to "Individual Commission Income Tax Rate 12%." Based on the rule matching results, the first ratio is set to determine whether a tax calculation rule and the to-be-matched tax calculation rule are in an update relationship, providing a clear decision-making basis for updating the tax calculation rule library. When the number of matching preset rule fields and their corresponding values ​​reaches or exceeds the first ratio, it indicates that the to-be-matched tax calculation rule and the existing tax calculation rule are highly consistent in core elements, with only some content changes. In this case, it should be considered an update to the existing rule. Generating rule update notifications clearly presents the specific content of rule changes to manual reviewers, allowing them to quickly understand the rules that need to be updated and the details of the changes, improving review efficiency and accuracy. This mechanism prevents minor rule adjustments from being mistakenly identified as new rules, ensuring the stability of the tax calculation rule library system and ensuring that the platform can promptly and accurately implement tax changes into tax calculation rules, maintaining the compliance and accuracy of tax calculation operations.

[0146] S706, otherwise, a new prompt is generated based on the tax calculation rules to be matched.

[0147] It can be understood that the rule addition prompt is a prompt information generated by the system based on the tax calculation rules to be matched when the matching ratio between the tax calculation rules and the tax calculation rules to be matched does not reach the first ratio. It is used to inform the manual reviewer that the tax calculation rules to be matched are brand new rules and need to be added in the tax calculation rule library. The prompt content contains the complete content and relevant instructions of the tax calculation rules to be matched. When the matching ratio between the tax calculation rules and the tax calculation rules to be matched does not reach the first ratio, it means that the tax calculation rules to be matched are significantly different from the existing tax calculation rules in core elements. It is not an update to the existing rules, but a brand new tax calculation rule. Generating a rule addition prompt at this time can promptly feedback this situation to the manual reviewer, ensure that the new rules will not be missed, and ensure that the tax calculation rule library can be continuously enriched and improved as taxes change. Through the rule addition prompt, manual reviewers can accurately understand the content of the rules that need to be added and strictly check during the review process.

[0148] S708: Send the tax calculation rules to be matched or the new rule prompts to manual review.

[0149] Manual review is the process by which individuals with tax expertise and social e-commerce experience review and judge system-generated rule updates and new additions. Reviewers must confirm the accuracy, rationality, and applicability of the matching tax rules and decide whether to approve the corresponding updates or additions to the tax rule base. While the system can initially process rule changes resulting from tax information announcements through rule matching and prompt generation, the complexity of tax policies and the diversity of business scenarios still require a manual review process for final approval. Reviewers leverage their expertise and experience to review and revise the system's processing results.

[0150] S710, update the tax calculation rule library based on the manual review results.

[0151] It can be understood that the manual review result refers to the reviewer's conclusion on whether to approve the update or addition of the tax calculation rule base, as well as any modification suggestions and opinions that may be included, after reviewing the rule update / addition prompt. Updating the tax calculation rule base based on the manual review result means modifying, adding, or deleting the rules in the tax calculation rule base according to the reviewer's approval, so as to ensure that the tax calculation rule base is consistent with the latest tax policies and business needs.

[0152] In one embodiment, the CPS commission settlement tax calculation method further includes: after the tax calculation rule library is updated, backtesting the updated tax calculation rules using historical tax calculation data. If the backtesting results do not meet the set requirements, a rule review prompt is issued based on the updated tax calculation rules.

[0153] Historical tax calculation data is the data generated by social e-commerce platforms during past tax calculation cycles, using the original tax calculation rules to calculate user CPS commissions. This data covers multiple dimensions, including user basic information, commission income, tax calculation basis, applicable tax rate, tax payable, and actual tax paid. Backtesting involves applying updated tax calculation rules to historical tax calculation data, simulating the tax calculation process, and comparing the simulated tax calculation results with the original tax calculation results to evaluate the rationality, accuracy, and applicability of the updated tax calculation rules. This process aims to verify the effectiveness of the new rules in historical scenarios and identify potential issues. New rules are inherently uncertain before they are implemented in practice. Backtesting with historical tax calculation data allows for simulation validation of updated tax calculation rules based on existing real-world business scenarios and tax calculation results. Historical tax calculation data encompasses tax calculation scenarios across a wide range of business scenarios and user types. By applying the updated rules to this data, we can comprehensively examine the tax calculation performance of the new rules under different conditions.

[0154] Set requirements are the standards and conditions pre-determined by social e-commerce platforms to evaluate the eligibility of backtesting results. These include, but are not limited to, the error range for tax calculation results (e.g., the absolute error in tax payable must not exceed a certain amount, or the relative error must not exceed a specific percentage), consistency requirements for tax calculation rule application (e.g., tax calculation results must remain consistent across the same business scenario), and the expected impact on user and platform interests. Rule review prompts are automatically generated by the system when backtesting results fail to meet these set requirements. These prompts include detailed information about the updated tax calculation rules, anomalies in the backtesting results, and analysis of potential issues. These prompts notify relevant personnel to review and re-examine the updated tax calculation rules. A prompt might read, for example, "The updated XX tax calculation rule exhibited an error of more than 10% in calculating the tax payable for XX type of user during backtesting. Please review the rule logic and parameter settings." Set requirements provide a clear basis for evaluating backtesting results. Failure to meet these set requirements indicates that the updated tax calculation rules exposed issues during historical scenario simulations, potentially due to errors in rule logic, improper parameter settings, or incomplete consideration of business scenarios. Issuing rule review prompts can promptly feed back these issues to tax professionals, system developers, or relevant auditors, prompting them to re-examine and review the updated tax calculation rules. Based on the prompt information, reviewers can conduct in-depth analysis of the causes of the problems, such as checking whether the tax rate calculation in the rules is correct and whether the judgment of tax calculation conditions is accurate, and make corresponding corrections and optimizations. This mechanism establishes an effective communication bridge between backtesting and manual review. By promptly discovering and resolving problems, it prevents flawed tax calculation rules from being applied to actual business, ensures the accuracy and compliance of tax calculation work, and protects the interests of the platform and users. It also helps to continuously improve the tax calculation rule base and enhance the reliability and stability of the platform's tax calculation system.

[0155] The present application provides a CPS commission settlement and tax calculation method device, including: a user portrait module, which is used to generate a user portrait based on the user information of the target user. A screening module, which is used to match the user portrait in the tax calculation rule library and filter out multiple tax calculation rules to be selected. A benefit determination module, which is used to evaluate the solution benefit of each tax calculation rule to be selected. A selection module, which is used to select the tax calculation rule to be selected with the highest solution benefit as the target tax calculation rule. A recommendation module, which is used to recommend the target tax calculation rule to the target user. A tax calculation module, which is used to calculate the tax of the CPS commission of the target user according to the target tax calculation rule in response to the target user's confirmation information for the target tax calculation rule.

[0156] For the specific limitations of the CPS commission settlement and tax calculation device, please refer to the limitations of the CPS commission settlement and tax calculation method above, which will not be repeated here. The various modules in the above-mentioned CPS commission settlement and tax calculation device can be implemented in whole or in part through software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0157] The present application provides a computer device comprising one or more processors and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the steps of the CPS commission settlement and tax calculation method in any of the above embodiments of the claims are executed.

[0158] Schematically, as Figure 8 As shown, Figure 8 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. Figure 8 Computer device 800 includes a processing component 802, which further includes one or more processors, and memory resources represented by memory 801 for storing instructions executable by processing component 802, such as application programs. The application programs stored in memory 801 may include one or more modules, each corresponding to a set of instructions. Furthermore, processing component 802 is configured to execute the instructions to perform the steps of the CPS commission settlement and tax calculation method according to any of the above-described embodiments.

[0159] The computer device 800 may further include a power supply component 803 configured to perform power management of the computer device 800 , a wired or wireless network interface 804 configured to connect the computer device 800 to a network, and an input / output (I / O) interface 805 .

[0160] The present application provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the CPS commission settlement and tax calculation method in any of the above embodiments.

[0161] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0162] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0163] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A CPS commission settlement and tax calculation method, characterized in that: include: Generate user profiles based on target users’ user information; Matching the user profile in the tax calculation rule library to screen out multiple tax calculation rules to be selected; Evaluate the benefits of each of the proposed tax calculation rules; Select the candidate tax calculation rule with the highest benefit of the solution as the target tax calculation rule; recommending the target tax calculation rule to the target user; In response to confirmation information of the target user regarding the target tax calculation rule, the CPS commission of the target user is taxed according to the target tax calculation rule.

2. The CPS commission settlement and tax calculation method according to claim 1, characterized in that: The user information includes user sub-information of multiple dimensions. Generating a user profile based on the user information of the target user includes: For the user sub-information of any dimension, determine the label mapping relationship corresponding to the dimension; According to the mapping relationship between the user sub-information and the tag, the user sub-information is mapped to the corresponding user tag to obtain the user portrait.

3. The CPS commission settlement and tax calculation method according to claim 2, characterized in that: The tax calculation rule library includes multiple tax calculation rules and their corresponding applicable attributes. Matching is performed in the tax calculation rule library based on the user profile to screen out multiple tax calculation rules to be selected, including: According to each of the user tags in the user portrait, matching is performed with the applicable attributes corresponding to each of the tax calculation rules to obtain a matching degree corresponding to each of the tax calculation rules; The matching rule with a degree greater than a matching degree threshold is determined as the tax calculation rule to be selected.

4. The CPS commission settlement and tax calculation method according to claim 3, characterized in that: Also includes: After evaluating the scheme benefits of each of the tax calculation rules to be selected, determining corresponding correlation coefficients according to the scheme benefits and the matching degrees of each of the tax calculation rules to be selected; Determining an average correlation coefficient based on the correlation coefficients; If the average correlation coefficient is lower than a preset correlation threshold, lowering the matching threshold; Otherwise, the matching threshold is increased.

5. The CPS commission settlement and tax calculation method according to claim 1, characterized in that: The evaluation of the benefits of each of the proposed tax calculation rules includes: Determining a user group corresponding to the target user according to the user portrait; Determine the user income corresponding to each of the to-be-selected tax calculation rules based on the mapping relationship between the tax rate levels corresponding to the user groups and the tax rates in each of the to-be-selected tax calculation rules; Determine the platform revenue corresponding to each of the tax calculation rules to be selected based on the average operating cost corresponding to each of the tax calculation rules to be selected; Based on the user income and the platform income, the plan income of each of the tax calculation rules to be selected is evaluated.

6. The CPS commission settlement and tax calculation method according to claim 1, characterized in that: The taxing of the CPS commission of the target user according to the target tax calculation rule includes: Summarize the withdrawal data of the target user during the tax calculation period; Determining tax calculation data based on the withdrawal data and the target tax calculation rules; Fill in the tax calculation data into the declaration form template to obtain the tax calculation declaration form; Upload the tax declaration form.

7. The CPS commission settlement and tax calculation method according to claim 6, characterized in that: Before uploading the tax return form, the following steps are also included: Determine whether the total income recorded in the tax declaration form is consistent with the target user's total withdrawal amount, whether the withholding tax recorded in the tax declaration form is consistent with the target user's actual deduction amount, and whether the total self-invoicing amount recorded in the tax declaration form is consistent with the actual withdrawal amount; If any inconsistency occurs, a manual screening prompt will be generated based on the inconsistent item, and the manual review process of the tax declaration form will be triggered based on the manual screening prompt.

8. The CPS commission settlement and tax calculation method according to claim 1, characterized in that: Also includes: Capture tax information announcements from the set data source according to the set crawling cycle; Determining whether the tax calculation rule library needs to be updated based on the tax information announcement; If so, the tax calculation rule library is updated according to the tax information announcement.

9. The CPS commission settlement and tax calculation method according to claim 8, characterized in that: The tax calculation rules in the tax calculation rule library include a plurality of preset rule fields. The determining whether the tax calculation rule library needs to be updated according to the tax information announcement includes: Inputting the tax information announcement and the preset rule field table into the target macro model to instruct the target macro model to determine whether the tax information announcement involves a change in the tax calculation rules, and if so, outputting the tax calculation rules to be matched; the preset rule field table includes all occurrences of the preset rule fields; If the target large model outputs the to-be-matched tax calculation rule, then the tax information announcement determines that the tax calculation rule library needs to be updated; Otherwise, it is determined that the tax calculation rule base does not need to be updated.

10. The CPS commission settlement and tax calculation method according to claim 9, characterized in that: The updating of the tax calculation rule library according to the tax information announcement includes: Matching the to-be-matched tax calculation rule with the tax calculation rule respectively; If the tax calculation rule and the to-be-matched tax calculation rule have a first ratio of the preset rule fields and their corresponding values ​​matched, then generating a rule update prompt based on the matched tax calculation rule and the to-be-matched tax calculation rule; Otherwise, a new prompt is generated based on the tax calculation rules to be matched; Sending the rule update prompt or the rule new addition prompt to manual review; The tax calculation rule library is updated according to the manual review results.

11. The CPS commission settlement and tax calculation method according to claim 8, characterized in that: Also includes: After the tax calculation rule library is updated, backtesting the updated tax calculation rules using historical tax calculation data; If the backtest results do not meet the set requirements, a rule review prompt will be issued based on the updated tax calculation rules.

12. A CPS commission settlement and tax calculation method device, characterized in that: include: User portrait module, used to generate user portraits based on the user information of target users; A screening module is used to match the user profile in the tax calculation rule library and screen out multiple tax calculation rules to be selected; A benefit determination module, for evaluating the benefits of each of the proposed tax calculation rules; A selection module is used to select the candidate tax calculation rule with the highest benefit of the solution as the target tax calculation rule; A recommendation module, configured to recommend the target tax calculation rule to the target user; The tax calculation module is used to calculate the tax on the CPS commission of the target user according to the target tax calculation rule in response to the confirmation information of the target user for the target tax calculation rule.

13. A computer device, characterized in that: The system comprises one or more processors and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the steps of the CPS commission settlement and tax calculation method according to any one of claims 1 to 11 are executed.

14. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to execute the steps of the CPS commission settlement and tax calculation method as described in any one of claims 1 to 11.

Images